The objective of this study is to assess determinants of local residential mobility and long distance migration among households headed by elderly men, with an emphasis on the variables that express the economic position of the individuals and families: income, net worth and housing characteristics. The overall hypothesis tested is that selected socioeconomic characteristics of individual and his household and characteristics of his housing are determinants of residential mobility and migration;The objectives are accomplished through a longitudinal analysis of the 1971, 1975 and 1979 waves of the Retirement History Study (a 10-year longitudinal study of retirement in the United States). The sample for this study is limited to 3906 households with a male head who were interviewed in 1971, 1975 and 1979;The main analyses consist of a series of logistic regressions to identify significant determinants of residential mobility and migration. Conditions in 1975, changes in selected socioeconomic variables between 1971 and 1975, and previous change of residence between 1971 and 1975 are used to predict residential mobility and migration between 1975 and 1979;Four different analyses form the core of the study. In the first, the dependent variable is any change of residence, including both residential mobility and migration. Then the dependent variables are respectively, residential mobility and migration. Finally, those who are residentially mobile are compared to the migrants;The overall hypothesis is partially supported by the analysis. The significant predictors of residential mobility are somewhat different from those of migration. Tenure remained the strongest, overriding determinant of both residential mobility and migration, followed by change of residence between 1971 and 1975. Beyond tenure and previous change of residence, the determinants are different, however. Health status and change of marital status are significant predictors of migration. Tenure and health status differentiate between the two types of moving behavior. The primary conclusion is that economic status does not affect either migration or residential mobility directly. Rather any effects are indirect, through tenure.